A Receiver-centric Approach to Interference Management: Fairness and Outage Optimization.

Effective interference management in the multiuser interference channel necessitates that the users form their transmission and interference management decisions in coordination, and adapt them to the state of the channel. Establishing such coordination, often facilitated through information exchange, is prohibitive in fast-varying channels, especially when the network size grows. This paper focuses on the multiuser Gaussian interference channel and offers a receiver-centric approach to interference management. In this approach, the transmitters deploy rate-splitting and superposition coding to generate their messages according to independent Gaussian codebooks. The receivers can freely decode any arbitrary set of interfering messages along with their designated messages in any desired joint or ordered fashion, and treat the rest of the interferers as Gaussian noise. The proposed receiver-centric interference management approach is applied to two class of problems (outage optimization and fairness-constrained rate allocation), and constructive proofs are provided to establish the following properties for the proposed approach: 1) the optimal set of codebooks to be decoded by each receiver is a local decision made by each receiver based on its local channel state information (CSI); 2) the globally optimal transmission rates are related to locally optimal rates computed by the receivers based on their local information, which implies that the transmitters do not require explicit knowledge of the CSI and can determine their rates via limited feedback from the receivers; and 3) obtaining the optimal interference management strategy at each receiver has controlled complexity. [ABSTRACT FROM AUTHOR]